cv

Algorithm Engineer at Ant Group since April 2026, researching pre-trained models for structured data and time series. Recent work includes FEAT, OneDB (SIGKDD 2026), and QBase (TKDE 2026). Explore my experience and research below, or download the PDF CV.

General information

Name Tang Qian
Label Algorithm Engineer at Ant Group
Email qt.tang.qian@zju.edu.cn
Phone +86-19982050103
Url https://qt.qiantech.online/
Summary Algorithm Engineer at Ant Group since April 2026, researching pre-trained models for structured data and time series.

Work

  • 2026.04 - Present
    Algorithm Engineer
    Ant Group
    Research on pre-trained models for structured data and time series.
    • Pre-trained structured data models
    • Pre-trained time-series models

Volunteer

  • 2023 - 2025

    Hangzhou, China

    Class Monitor
    Software Engineering Class 6, Grade 2023
    Serving as the Class Monitor for Software Engineering Class 6, Grade 2023.
  • 2023 - 2024

    Hangzhou, China

    President of Student Union
    College of Computer Science and Technology, Zhejiang University
    Serving as the President of Student Union for the College of Computer Science and Technology.
  • 2021 - 2022

    Chengdu, China

    President of the College Student Union
    Southwest Jiaotong University
    Served as President of the College Student Union.
  • 2021 - 2023

    Chengdu, China

    Deputy Secretary
    College League Committee, Southwest Jiaotong University
    Served as Deputy Secretary of the College League Committee.
  • 2020 - 2023

    Chengdu, China

    Class Monitor
    Computer Science Class 1, Grade 2019
    Served as the Class Monitor for Computer Science Class 1, Grade 2019.

Education

  • 2023.09 - Present

    Hangzhou, China

    Master's
    Zhejiang University
    Computer Science and Technology
    • Vector Search
    • Distributed Systems
    • Big Data Analysis
  • 2019.09 - 2023.06

    Chengdu, China

    Bachelor's
    Southwest Jiaotong University
    Computer Science and Technology
    • Spatio-temporal data analysis
    • Graph neural networks

Publications

Projects

  • FEAT: A Linear-Complexity Foundation Model for Extremely Large Structured Data
    A linear-complexity foundation model for extremely large structured data, supporting zero-shot classification, regression, and missing-value imputation.
    • Structured data foundation models
    • Bidirectional state-space modeling and gated linear attention
    • Linear sample-axis complexity
    • Zero-shot learning
  • 2025 - Present
    High-Dimensional Data Retrieval Systems
    Research project exploring approximation algorithms to improve query throughput in distributed nearest neighbor search, developing techniques to handle data skew and load balancing in graph-based distributed indexes, and conducting benchmark evaluation against state-of-the-art vector search systems.
    • Approximation Algorithms
    • Distributed Nearest Neighbor Search
    • Data Skew Handling
    • Load Balancing
    • Benchmark Evaluation
  • 2023 - 2024
    Distributed Indexing for Metric Space
    Designed and implemented a scalable distributed architecture using Spark for processing large-scale metric datasets, optimized data partitioning algorithms to reduce network communication overhead by 40% while maintaining search accuracy, and conducted extensive experimental evaluation.
    • Distributed Architecture
    • Spark
    • Data Partitioning Optimization
    • Network Communication Reduction
    • Experimental Evaluation
  • 2024 - 2025
    OneDB: Multi-Modal Data Similarity Search Framework
    Developed a distributed framework for multi-modal data similarity search supporting efficient querying across heterogeneous data types, designed a dual-layer indexing strategy, implemented a lightweight weight learning model, and achieved significant performance improvements over state-of-the-art solutions.
    • Multi-Modal Data
    • Distributed Framework
    • Heterogeneous Data Types
    • Dual-Layer Indexing
    • Weight Learning Model
  • 2024 - 2025
    QBase: A Query Aware Vector Database System
    Development of QBase, an innovative query-aware vector database system designed to optimize vector search with dynamic query optimization, implementing cost-based query planning algorithms to adaptively select optimal index structures.
    • Vector Database System
    • Dynamic Query Optimization
    • Cost-based Query Planning
    • Adaptive Index Selection
  • 2021 - 2022
    Graph-based Traffic Analysis Platform
    Built data preprocessing pipeline for cleaning and normalizing traffic sensor data from multiple urban sources, implemented spatio-temporal graph neural network models, and designed interactive visualization dashboard to present traffic patterns and prediction results.
    • Data Preprocessing
    • Traffic Sensor Data
    • Spatio-Temporal Graph Neural Networks
    • Python
    • PyTorch
    • Interactive Visualization

Awards

Skills

Programming Languages
C++
Scala
Python
Frameworks & Tools
Spark
Hadoop
Distributed Computing

Languages

Chinese
Native speaker
English
Professional working proficiency

Interests

Database Systems
Vector Search
High-Dimensional Data
Similarity Search
Distributed Databases